Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports Community BlogPrivacy & TermsContact
    © Neuronpedia 2025
    Privacy & TermsBlogGitHubSlackTwitterContact
    1. Home
    2. Apollo Research · Taylor · Identifying Functionally Important Features with End-to-End Sparse Dictionary Learning
    3. GPT2-Small
    4. Residual Similar CE E2E Recon
    5. 6-RES_SCEFR-AJT
    6. 307
    Prev
    Next
    INDEX
    Explanations

    programming code structure and syntax patterns

    oai_token-act-pair · gpt-3.5-turbo
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    neuronpedia/gpt2-small__res_scefr-ajt/6-res_scefr-ajt
    Prompts (Dashboard)
    12,288 prompts, 128 tokens each
    Dataset (Dashboard)
    Skylion007/openwebtext
    Features
    46,080
    Data Type
    torch.float32
    Hook Point
    blocks.6.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    apollo-research/Skylion007-openwebtext-tokenizer-gpt2
    Hook Point Layer
    6
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Neuron Alignment
    Index
    Value
    % of L₁
    655
    +0.13
    0.5%
    9
    +0.12
    0.4%
    128
    +0.11
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    6
    +0.13
    0.05
    373
    +0.12
    -0.03
    326
    +0.11
    0.05
    Negative Logits
     cones
    -0.95
     cone
    -0.92
     shocks
    -0.88
    esome
    -0.85
     restraints
    -0.85
    ulative
    -0.85
     timeless
    -0.84
     charms
    -0.83
     foss
    -0.82
     manif
    -0.81
    POSITIVE LOGITS
    @#&
    1.36
    important
    1.34
    ctor
    1.07
    RFC
    1.01
    #$
    0.99
    OPS
    0.95
    ailable
    0.95
    ECH
    0.94
    empty
    0.94
    include
    0.91
    Activations Density 0.551%

    No Known Activations